The terms, defined against each other
Most confusion in this area is four adjacent concepts collapsed into one word.
- No-show rate
- Appointments where the patient did not attend and gave no notice, as a share of appointments scheduled. The words no notice are what separate it from everything below, and they are what most spreadsheets quietly omit.
- Cancellation rate
- Appointments cancelled with enough notice to be refilled. This is not a loss and should not be counted as one — a practice with a high cancellation rate and a low no-show rate has patients who communicate, which is the outcome you want.
- Late cancellation
- Cancelled inside a window — commonly 24 or 48 hours — too late to resell the slot. Economically identical to a no-show and psychologically completely different, which is why practices file it separately and then forget to add it back.
- Effective loss rate
- No-shows plus late cancellations, minus whatever was genuinely refilled. Not a standard term, and the only one of the four that maps to money. If you are going to track a single number, track this.
- Attendance rate
- The complement, used mainly in public health reporting: attended appointments over scheduled. Reported this way because a rising number reads better in a board pack than a falling one.
Also called: missed appointment rate · DNA rate · failed-to-attend rate
See also: Cancellation rate, Late cancellation
Also called: advance cancellation rate
See also: No-show rate
Also called: short-notice cancellation
See also: No-show rate, Effective loss rate
See also: No-show rate, Late cancellation
Also called: show rate
See also: No-show rate
Why comparisons mislead
Three practices can each report 12% while running very different businesses. One counts every booked slot including the ones staff cancelled. One counts only patient-caused absences. One counts late cancellations as no-shows because the money is the same.
The denominator moves as much as the numerator. A practice that books six weeks ahead has more time for life to interfere than one that books three days ahead, and both will describe their rate as if it were a measure of their patients rather than of their scheduling horizon.
The practical consequence: your own trend over time is informative, and your position against a published benchmark mostly is not.
Published ranges by sector
The spread rather than the midpoint, because the spread is the finding. Where sources disagree, the wider range is shown.
| Sector | Reported range | Typical driver of the high end |
|---|---|---|
| General medical practice | 5–15% | Same-day and short-horizon booking, high volume |
| Hospital outpatient clinics | 10–30% | Long waits between referral and appointment |
| Dental practice | 8–20% | Recall appointments booked six months ahead |
| Mental health services | 15–30% | The condition itself is frequently the reason for the absence |
| Physiotherapy and rehab | 10–20% | Multi-session courses where later sessions feel optional |
| Salons and personal services | 10–20% | Low switching cost, no clinical relationship |
| Specialist private consultations | 3–10% | A fee already paid, and a scarce appointment |
Source: Ranges compiled from published health-service statistics and sector studies, August 2026. These are ranges reported across many studies of varying quality, not a meta-analysis, and they are not comparable to each other because the underlying definitions differ.
The mental health row is the one that deserves care. A high no-show rate there is a symptom, not a discipline problem, and reminder programmes aimed at it should be designed by someone who understands that. This product is not built for that setting and this page is not going to imply it is.
Measuring yours so the number means something
Four steps, one afternoon, and afterwards you have a figure that is comparable with itself over time.
- 1
Pick a definition and write it down
The recommended one: patient did not attend and gave less than 24 hours' notice, over all patient-facing slots scheduled.
Writing it down matters more than which one you pick, because the entire value is comparability across months and staff.
- 2
Separate the three buckets from the start
No-show, late cancellation, advance cancellation. Three columns, filled in as it happens.
Reconstructing this later is impossible. Nobody remembers in March whether the January absence came with a phone call.
- 3
Segment by appointment type before you segment by anything else
First visits, follow-ups, and recall appointments behave completely differently, and a single blended figure hides where the problem actually is.
The usual finding is that one category carries most of the loss. Practices that discover this stop trying to fix the whole schedule and fix one clinic session.
- 4
Track it monthly, look at it quarterly
Monthly is noisy at small volume. A twelve-appointment week can swing the rate by five points on one absence.
The mistake is reacting to monthly movement. At 60 appointments a week you need about a quarter before a change is distinguishable from noise.
None of this needs software. A spreadsheet with four columns does it, and doing it in a spreadsheet for a quarter is the cheapest way to find out whether the problem is big enough to buy anything for.
Questions
What is a good no-show rate?
Should late cancellations count?
Does charging a no-show fee reduce the rate?
Why do hospital clinics report such high rates?
Tell me when this exists
In the context field: your rate and how you define it. The definitions are the interesting part.